Asian Journal

Asian Journal of
of Innovation
Innovation and
and Policy
Policy (2018)
(2018) 7.1:055-078
7.1:055-078
DOI: http//dx.doi.org/10.7545/ajip.2018.7.1.055

Entrepreneurial Learning and Indian Tech Startup

Survival: An Empirical Investigation
H S Krishna*

Abstract This paper investigates the linkage between the mode of transformation of
entrepreneurial learning into outcomes and the subsequent impact of these learning
outcomes in enhancing the survival of high-tech startups in India. The study uses data
from 45 high-tech startups headquartered across different locations in India for the
purpose of analysis. Survival Analysis of the data is conducted to determine which
mode of learning transformation and what type of en trepreneurial decision making
preference have a significant influence on the survival of Indian high-tech startups and
to what extent do they impact their survival. The results indicate that entrepreneur's
prior startup experience, explorative mode of learning transformation, causal decision
making of the entrepreneur and availability of funding for the startup as the key factors
that reduce the time to survival of Indian high-tech startups. They also provide key
insights on how these factors impact the startup survival in this region.

Keywords Entrepreneurial learning, high-tech startup survival, effectuation, India

I. Introduction

In recent times, high-tech startups have gained increased attention across the
world from multiple stakeholders in our society. Policy makers and
governments view these high-tech startups as a new way to realize the goals of
job creation, innovation and economic development (Kirchoff and Spencer,
2008). Young skilled individuals joining the workforce view startups as a
preferred occupational choice. High-tech startups have been extensively
studied in the entrepreneurship literature from multiple discipline-based
perspectives, the prominent ones being economic, strategic management,
evolutionary and behavioral sciences.
While high technology startup firms have been credited with contributing to
economic growth by way of job creation and innovation (Kirchhoff, 1994;
2008), a review of the characteristics of these startups reveal that they have a

Submitted, August 27, 2017; 1st Revised, January 23, 2018; Accepted, February 27, 2018
* Department of Management Studies, Indian Institute of Science, Bangalore-560012;

krishnah@iisc.ac.in

55
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

high failure rate (Stinchcombe, 1965). Given these seemingly contradicting

observations, one can infer that the contribution to economic development is
through those startup firms that have been able to overcome the difficulties
during the early stages of the firm lifecycle and emerged successfully. Hence,
high-tech startup survival has been one of the most probed phenomena from
different perspectives.
A couple of decades ago, entrepreneurship researchers focused on explaining
the phenomenon of startup lifecycle events such as survival and success from a
static perspective. However, of late, there has been broad consensus in the
community that entrepreneurship is a complex and dynamic process (Wiklund
et al., 2009; Kim and Shin 2017; Song et al., 2017). More specifically,
researchers have moved away from trying to identify traits of successful
entrepreneurs or from examining the characteristics of successful firms to a
more process-based mode of examining entrepreneurship. Recent studies have
attempted to understand entrepreneurship as “a continuous learning process”
(Politis, 2008). Researchers have applied the core concepts of learning theory
to different entrepreneurial contexts. Politis and Gabrielsson (2005) observed,
“Entrepreneurship occurs at the intersection of the individual’s perception of
an entrepreneurial opportunity and his or her ability to pursue that opportunity”.
Shane and Venkataraman (2000) further explained that entrepreneurs develop
different skills and capabilities and therefore keep piling up their stocks of
information as part of their careers. Furthermore, these stocks of information
gathered by the entrepreneurs influence their ability to recognize
entrepreneurial opportunities, act on them and exploit the situation to their
economic benefit.
Research from the past decade has delved deeper into the above facets,
trying to unravel how entrepreneurs learn from experiences, and how the
experiences further help the entrepreneur in enhancing the performance of the
venture or in reducing the time to survival of the startup. The role of prior
startup experience in entrepreneurial learning (Politis and Gabrielsson, 2005),
the distinction between entrepreneurial learning and entrepreneurial knowledge
(Reuber et al., 1990), the exploration of the intermediate process of
entrepreneurial learning where “experience is transformed into knowledge”
(Minniti and Bygrave, 2001; Politis and Gabrielsson, 2005) are some of the
notable contributions that have expanded our understanding in this domain.
Barring the above contributions, there are very limited studies that try to
integrate the above concepts and understand the variance in entrepreneur
learning, as well as the variance in learning outcomes that may impact the
survival of startups. This study is an initial attempt to understand the end-to-
end perspective of the causes of variance in entrepreneurial learning and
learning outcomes, and to study the impact of these variations on the survival
of the entrepreneur’s venture.

56
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Based on the aspects mentioned in the above discussion, the primary goal of
this study is to examine the linkage between entrepreneurial learning, its
modes of transformation to learning outcomes, and subsequently, the impact of
such learning outcomes on high-tech startup survival. To achieve this aim, we
first study different theoretical contexts that have enhanced our understanding
of entrepreneurial learning (Kolb, 1984; March, 1991). We then explore the
literature that describes the modes by which the entrepreneurial learning is
transformed into entrepreneurial knowledge (Politis and Gabrielsson, 2005;
Politis, 2008). Second, we identify one such learning outcome candidate - the
decision-making preference of the entrepreneur to examine if the mode of
learning transformation has an impact on learning outcomes, and whether this
in turn would impact the time to survival of high-tech startups (Sarasvathy,
2001; 2008). At the outset, we are interested to find out which is the dominant
learning transformation mode of entrepreneurs. And then, we seek to
understand if any particular preference for a decision-making style helps the
entrepreneurs to reduce the time to survival.
The reminder of the study is organized as follows. The next section presents
a review of literature, wherein a synthesis of past literature pertaining to
survival of startups and entrepreneurial learning processes is presented. The
literature review is then followed by a section that outline the conceptual
framework linking the process of entrepreneurial learning to its outcomes, and
subsequently the impact of learning outcomes to startup survival. Next is a
section that describes the research methodology, wherein the sample frame and
measures and variables used in the study, and the empirical methods of
analysis adopted. We then present the detailed analysis of the results obtained.
The study concludes with a discussion of the findings, as well as a summary of
the contributions made together with its limitations.

II. Literature Review

We present here the review of literature on entrepreneurial learning and

modes of transformation of learning into learning outcome. We then focus on
examining the existing literature that deals with entrepreneurial decision-
making as one such learning outcome. Finally, we explore the literature linking
such learning outcomes to high-tech startup survival.

1. Entrepreneurial Learning

In recent years, researchers have applied the concepts and ideas of

experiential learning theory to the field of entrepreneurship in an attempt to

57
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

extend existing knowledge on entrepreneurial learning. These contributions

observed that entrepreneurs learn through a hands-on and practical process
where entrepreneurs develop and accumulate knowledge by applying their
existing skills and knowledge in new ventures, thereby developing new
knowledge in the process (Sullivan, 2000; Rae and Carswell, 2001; Politis and
Gabrielsson, 2005; Middleton and Donnellon, 2014).
Since there are many different contexts in which entrepreneurial learning can
be studied, it is natural to expect multiple definitions of the same. In their
meta-analysis review, Wang and Chugh (2014) summarized a few well-
accepted definitions of entrepreneurial learning. Entrepreneurial learning has
been defined as “learning in the entrepreneurial process” (Politis, 2005). Rae
(2005) defined entrepreneurial learning as, “learning to recognize and act on
opportunities, and interacting socially to initiate, organize and manage
ventures”. Young and Sexton (2003) defined entrepreneurial learning as ”the
variety of experiential and cognitive processes used to acquire, retain and use
entrepreneurial knowledge”.
Two theories of entrepreneurial learning, namely, experiential learning
(Cope, 2003; Minniti and Bygrave, 2001) and organizational learning (Covin
et al., 2016; Covin et al., 2006; Wang, 2008), dominate the literature related to
entrepreneurial learning. The experiential learning contributions can primarily
be traced back to the work of Kolb (1984). Subsequently, many variants and
related theories have emerged inspired by Kolb’s contributions. Key among
them are “learning-by-doing” (Balasubramanian, 2011; Cope, 2003), “learning
from past business experience” (Lamont, 1972), “learning from positive and
negative experiences” (Minniti and Bygrave, 2001), “learning from past
experience” (Sardana and Scott-Kemmis, 2010), and “learning from
participation and from the experience of others who are involved in startup
related activities” (Lévesque et al., 2009).
There have also been significant contributions emanating from the theories
of organizational learning, the origins of which could be traced to March
(1991). Some of the key contributions that have originated from this strand of
theory are “single- and double-loop learning” (Argyris and Schön, 1978),
“organizational learning” (Huber, 1991), “and absorptive capacity and external
learning” (Cohen and Levinthal, 1990). The theory of organizational learning
has been applied to entrepreneurial learning studies in multiple contexts by
different scholars. For example, Chaston et al. (2001) examined whether
entrepreneurial firms used higher-order learning. Lant and Mezias (1990)
studied on aspects of organizational learning theory that helped to concep-
tualize entrepreneurship.

58
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

2. Modes of Entrepreneurial Learning Transformation

With the above background and overview of entrepreneurial learning, we

now shift our focus to understand how this learning gets transformed into
knowledge, which the entrepreneur uses as part of his daily activities to
perform his duties. Prior studies that examined entrepreneurial learning as an
experiential learning process have stressed that the experience of the
entrepreneur (prior skills and knowledge) and the new knowledge acquired by
the entrepreneur need to be examined separately (Reuber and Fischer, 1994;
Politis 2005; Mian et al., 2016).
In an attempt to provide linkage between the experiences of entrepreneurs
and the knowledge acquired by entrepreneurs, a few studies have explored the
transformation process between the above two aspects (Reuber and Fischer,
1999). Politis and Gabrielsson (2005) identified two distinctive courses of
transformation of entrepreneur’s experiences into knowledge, one through
exploitation and the other through exploration. They further explained, “in the
process of exploitation, entrepreneurs choose actions that replicate or are
closely related to the ones they have previously taken, thereby exploiting their
pre-existing knowledge”. In the latter case, that of exploration, “the
entrepreneurs choose entirely new actions that are distinct from the ones that
they have already taken, thereby exploring new domains where they have no
previous experience”.
Thus, the exploratory transformation process can be viewed as “variance-
seeking learning” that increases “performance variance”, whereas the ex-
ploitative transformation process as “mean-seeking learning” that improves
“mean performance and decreases variance” (McGrath, 2001). Middleton and
Donnellon (2014) studied the entrepreneurs in accelerators and incubators and
noted that entrepreneurs learnt the what, why and how of the execution of
entrepreneurial actions at these entities. However, it has to be noted that these
modes of transformation are not mutually exclusive. Levinthal and March
(1993) posited that a proper balance between exploration and exploitation is
crucial for ensuring survival and success of ventures. Mintzberg and Waters
(1982) had drawn attention to the need of this balance by stating that
“exploitation of successful new ideas provides resources to support new
exploration”. These studies indicate that entrepreneurs need to tread the fine
line of balancing between risk and returns for every action they undertake.

3. Decision Making as a Learning Outcome

The discussion thus far has elaborated on the importance of entrepreneurial

learning, as well as the need to appreciate the transformation of experiences

59
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

into knowledge. Continuing with the same perspective, we now seek to

understand how entrepreneurs used this knowledge to carry out various actions
and activities. This examination will provide a measure of the learning and
utilization of knowledge that they have developed. Politis (2008) identified
three learning outcomes that entrepreneurs can be measured against as they
embarked on the activities of managing a new venture.
Decision-making is one such critical learning outcome that can be used to
measure the effectiveness of the entrepreneurial journey. Gabrielsson and
Politis (2011) noted that entrepreneurs, specifically during the initial phases of
the startup lifecycle, have to take many decisions on a daily basis, such as fine-
tuning their business idea, identifying or carving out a niche market for their
offerings, addressing technical problems, obtaining the required resources,
hiring key personnel, etc. They observed that all these decisions have a long-
term impact (Boeker, 1988) and crucial and important decisions taken at the
early lifecycle stages may have long-lasting impact on the future success and
growth of the new venture (Vohora et al., 2004). Therefore, studying
entrepreneurial decision-making as a learning outcome is useful in aiding a
better appreciation of the process whereby entrepreneurs create economic
value by way of exploitation of the entrepreneurial opportunity.
Sarasvathy (2001) identified two modes of reasoning that were used by
entrepreneurs when they made decisions on their business or new ventures:
causation and effectuation. Causation is described as “a problem-solving
decision-model that rests on the logic of prediction” (Sarasvathy, 2001;
Sarasvathy and Dew 2005). In this model, the entrepreneur makes a decision
based on a choice among the available means to achieve a given result. The
choice of means in this case is motivated by the entrepreneur’s knowledge
along with the final result that the entrepreneur wants to create (Gabrielsson
and Politis, 2011). The core driving logic of this decision-making model is that
“the entrepreneur can control the future, to the extent that he can predict the
future”. This logic is accomplished by continuous planning, collection of
information to understand and analyze the progress of the activities against the
plan, and identifying the root cause of deviations discovered for those activities
that had different outcomes against the plan (Gabrielsson and Politis, 2011).
Effectuation is based on a different problem-solving decision model that
relies on the logic of “control”, where the entrepreneur pursues entrepreneurial
activities based on what can be done, given the means that are currently
available (Sarasvathy, 2001; Sarasvathy and Dew 2005). In this model, the
entrepreneur makes a choice of the end result that can be created, with the
given means. The choice of the final result achieved is dependent on the
entrepreneur. The choice of this end result is driven by the entrepreneur’s
ability to discover and use contingencies. The core logic driving the effectual
model is that, to the extent that an entrepreneur can control the future, he does

60
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

not need to predict it (Sarasvathy, 2001). While effectuation considers the

external environment to be endogenous to the entrepreneur’s decisions and
actions, causation views the external environment as an exogenous factor
influencing the entrepreneurial decision-making (Gabrielsson and Politis,
2011).
Although causation and effectuation seem to be describing two com-
plementary approaches to entrepreneurial decision-making, Sarasvathy (2001)
underlined that “both decision-making logics are integral parts of human
reasoning and can occur simultaneously, overlapping and intertwining over
different contexts of decisions and actions”. Gabrielsson and Politis (2011)
noted that entrepreneurs prefer and use either effectual or causal reasoning or
both at different times, depending on their individual circumstances and
preferences. Although the decision-making preference of entrepreneurs would
be heavily influenced by the unique situational context (Douglas, 2005), in this
study the primary focus is on examining whether there is entrepreneurial
preference for one type of decision-making logic over the other. Further, if
there is such preference, is this a result of a predominant style of learning
transformation, and finally, if this preference has any impact on the firm
survival. This approach is similar to studies conducted by Wiltbank et al. (2009)
and Brettel et al. (2012), where the former contrasted effectuation against
causation to examine angel investing outcomes; and the latter contrasted
effectuation against causation to examine decision making in the corporate
environment.

4. Survival of High-Tech Startups

From the lens of experiential learning, a couple of key abilities of

entrepreneurs are recognized as important factors in enabling entrepreneurs to
solve the challenges during the startups’ emergence and survival. These are
“abilities to recognize and act on entrepreneurial activities” (Shane and
Venkataraman, 2000) and “coping with liabilities of newness” (Stinchcombe,
1965; Shepherd et al., 2000). Gabrielsson and Politis (2011) indicated that the
capability to identify and work on entrepreneurial opportunities is usually a
mark of the successful entrepreneur.
Prior literature stresses that entrepreneurs need to have the right skills or past
experience to deal with the challenges of liability of newness that a new
venture presents. Politis (2008) observed that these liabilities are routine and
conventional problems related to the people and task management in the new
venture, the uncertainty that exists in the discovery of the value of new
product/service in the marketplace, problems with obtaining long-term external
finance, and challenges in hiring skilled and resourceful personnel to the new

61
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

organization. Prior studies have indicated that experienced entrepreneurs

would have acquired intangible knowledge about all the key stakeholders that
they need to deal with, as they start a new venture. They would have
established good relationships with reliable suppliers, have a deep insight on
the viable markets, availability of product and resources, which increases their
ability to identify and act on entrepreneurial opportunities (Covin et al., 2016;
Ronstadt, 1988; Shepherd et al., 2000). This body of literature therefore
suggests that habitual entrepreneurs, by virtue of their prior startup experience,
possess a greater ability to tolerate and withstand unfavorable shocks and to
take corrective actions as required in their new venture.
From a learning perspective, irrespective of whether the entrepreneur is
experienced or novice, it is pertinent to note that, as the entrepreneur immerses
himself in the process of setting up a new venture, he would need to
continually upgrade and enhance the critical resources and capabilities (Brush
et al., 2001). The outcome of this endeavor is largely determined by the
convergence of the entrepreneur’s skills, preferences and attitudes in response
to the various challenges they face (Markman and Baron, 2003). Rerup (2005)
observed that these experientially acquired capabilities would significantly
improve venture performance, when these activities take place in a favorable
external environment.

III. Conceptual Framework and Research Objectives

While the previous section dealt with the key theoretical issues concerning
entrepreneurial learning, its mode of transformation into knowledge, and its
role in impacting survival of high-tech startups, in this section, we seek to
provide the linkage to these identified tenets by way of developing a
conceptual framework. We detail the findings in subsequent sections.
From the above discussions, it is clear that entrepreneurs gain new
experiences and therefore develop new knowledge as an ongoing process
(Politis and Gabrielsson, 2005). Subsequent discussions have revealed that
exploration and exploitation are two dominant modes of transformation of
these new experiences to knowledge (Minniti and Bygrave, 2001). There have
also been studies that have identified key learning outcomes, which are a
demonstration of application of the acquired entrepreneurial knowledge
(Politis, 2008). One such important learning outcome is decision-making of the
entrepreneur that has been observed to make an impact on the survival or
success of the startup. Causation and effectuation have been discussed as two
such dominant decision-making styles that entrepreneurs employ for decision-
making (Sarasvathy, 2001).

62
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Building on these attempts, a conceptual framework is developed, which is

depicted in Figure 1, to answer the variance in entrepreneurs’ knowledge
acquisition and to link this variance to startup survival.

Figure 1 Conceptual framework of the study

This framework seeks to understand on how learning of entrepreneurs varies

in the process of learning transformation either through exploration or
exploitation. Second, it further tries to link the effect of variance of
entrepreneurial learning outcome to survival of startups. Based on the above
framework, the research objectives of this study are as follows:

- What is the dominant mode of learning transformation exhibited by the

early stage high-tech startup entrepreneurs?
- What is the dominant mode of decision-making preference exhibited by
early stage high-tech startup entrepreneurs?
- How does the dominant mode of learning and decision making pre-
ference affect the time to survival of high-tech startups?

The next section describes the data collected to validate the above
framework, the characteristics of the sample, the variables and measures used
in the study, the method of statistical analysis to validate the data.

IV. Scope, Data Description and Research Methods

1. Scope

The above objectives are studied with respect to high-tech startups

headquartered and operating across different cities in India. The scope of this
study was restricted to IT high-tech startups that have their registered
headquarters in India. Since there is no single credible database of startups
operating out of India, the database of high-tech startups was procured from
Industry associations such as National Association for Software and Services
Companies (NASSCOM) and Indian Software Product Industry Round Table

63
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

(iSPIRT), and Incubation Centers such as Microsoft Accelerator, T-Labs, N S

Raghavan Center for Entrepreneurial Learning at IIM Bangalore. This enabled
us to develop a consolidated database of high-tech startups. This consolidated
database contains data of 1567 startups.
To ensure that the consolidated database is representative of the overall
population, we obtained the demographic distribution of startups data from the
consolidated database, and compared it with well-accepted Industry Reports
such as the iSPIRT Product Industry Monitor Report 2014. We find that the
consolidated database is representative on all dimensions that we could assess:
industry demographics such as location, age of the startup, market sector,
founder profiles such as education, experience and external funding status.
The founders of 200 startups from this consolidated database were picked
and contacted with a request to participate in the study. These 200 startups
were chosen using a stratified sampling technique to ensure that they were
representative of the population. Among them, 72 founders responded to the
request, of which 27 founders did not want to participate in the study. The
research instrument was then sent to the remaining 45 founders who agreed to
participate in the study, indicating a 22.5% response rate. A semi-structured
interview schedule was personally administered by the researcher or
administered over telephone to all the founders who agreed to participate in the
study during the month of February 2014. The methodology followed for the
interview is described in the following sections.

2. Sample Characteristics

The duration of operation of the startups in the sample ranges from 6 months
to 120 months. About 73% of the startups in the sample had found their
product market fit (milestone for survival), while the remainder 27% were new
firms created, but yet to claim survival. About 91% of the founders in the
sample had a minimum of one-year industry experience. About 49% founders
of the sample had prior startup experience. About 94% of the founders
possessed either startup experience or industry experience at the time of
creation of their startup. The founders’ age at the time of starting up in the
sample ranged between 22 years to 49 years. In terms of founders’ education,
about 9% of the founders had a non-engineering graduate education (science,
commerce, arts degree), 44% of them had an engineering bachelor’s degree
and 47% had masters’ engineering degree or higher educational qualification
when they started their venture.

64
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

3. Variables and Measures

A semi-structured interview schedule was developed to collect the primary

data to analyze the objectives of this study. The schedule was designed to
collect basic profile information related to the founder and the startup, as well
as to collect information related to modes of learning transformation,
preference of decision-making styles and the time taken by the startup to reach
the survival milestone. Standard scales of measurement from prior literature
were adopted to collect information on learning transformation modes and
preference of decision-making styles. Industry experts and startup
entrepreneurs validated the schedule. The various dependent and independent
variables used in the study are as follows:

Dependent Variable The dependent variable used in this study is time to

survival of the startup in months. The time in months of operation of the
startup since formal incorporation (labelled as ‘stime’ for the analysis) and
whether the startup has achieved survival or not (labelled as ‘survst’ for the
analysis) taken together form the dependent variable. The respondents to our
questionnaire reported the month and year that they formally created the
startup. The startup that had achieved product-market fit at the time of
conducting the survey is considered to have survived. Startups that had not yet
achieved this milestone are considered not having survived at the time of
observation.
The event of product-market fit is used as a proxy to measure startup
survival. This milestone indicates that the startup has been able to achieve
repeatable sales with a standardized offering, and that it has now a formidable
set of initial customers that have validated the offering and are paying for the
offering.

Independent Variables: Mode of Learning Transformation The preference

between the two modes of learning transformation (exploratory, exploitative)
form one set of independent variables. This multi-item construct is adopted
from Politis and Gabrielsson (2005). This ten-item question provides the res-
pondents with a choice between exploration and exploitation using a Likert
scale consisting of five items each. The raw scores collected from the
interview schedule are converted to binary scale using weighted average
method and used for the survival analysis. This variable is labelled as ‘Ol’ for
the analysis.

Preference for Decision-Making The preference between two dominant

modes of decision-making (effectuation and causation) form the other set of

65
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

independent variables. This multi-item measure is borrowed from Brettel et al.

(2012). This scale contrasts effectuation items with causation items using a six-
point Likert scale to measure whether there is preference for an effectual or
causal approach of decision-making. This scale has a total of 23 items covering
four dimensions, namely, means versus goals (7 items), affordable loss versus
expected returns (5 items), partnerships versus competitive market analysis (4
items) and acknowledge the unexpected versus overcome the unexpected (6
items). The raw scores collected from the interview schedule are converted to
binary scale using weighted average method. This variable is labelled as
‘dlogic’ for the analysis.

Control Variables This study uses entrepreneur-specific and startup-specific

factors as control variables. Relevant industry and startup experience, age and
education of the entrepreneur represent the entrepreneur-specific factors that
are used as controls. The sales turnover in Indian currency, the research and
development capability and funding status of the startup form the startup-
specific factors that are controlled for.

Relevant Industry Experience A discrete dichotomous variable that indicates

whether or not the founder has previous industry experience has been used for
analysis. This variable takes the value of 1 for every startup founder who has
industry experience, prior to founding the current startup considered for the
study. A value of 0 for this variable indicates that the founder of the startup
does not possess any previous industry working experience. This variable is
labelled as ‘fiexp’ for the analysis.

Prior Startup Experience A discrete dichotomous variable that indicates

whether or not the founder has prior startup experience has been used for
analysis. This variable takes the value of 1 for every startup founder who has
experience working in a startup either as an employee or as a founder, prior to
founding the current startup considered for the study. A value of 0 for this
variable indicates that the founder of the startup does not possess any previous
startup experience. This variable is labelled as ‘fsexp’ for the analysis.

Age of the Entrepreneur The age of the entrepreneur in years, at the time of
founding the current startup has been used for analysis. This variable is
labelled as ‘fage’ for the analysis.

Education of the Entrepreneur The education of the entrepreneur is

categorized using two dummy variables. The base reference variable indicates
graduate education without an engineering degree (degree in Science, Arts and
Others); the first dummy variable indicates graduate education with a technical

66
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

(engineering) degree, and the second dummy variable indicating education

with a technical master’s degree or above. This variable is labelled as ‘fedn’
for the analysis. The base reference variable takes the value of 1 for every
founder of the startup whose education credentials is a non-engineering degree.
A value of 0 for the base reference variable indicates the absence of a non-
engineering degree of the founder. The first dummy variable ‘fedn(1)’ takes
the value of 1 for every startup founder who has a technical (engineering)
degree as his/her education credentials. A value of 0 for this variable indicates
the absence of a technical (engineering) degree of the founder. The second
dummy variable ‘fedn(2)’ takes the value of 1 for every startup founder who
has education credentials of a master’s technical (engineering) degree or above.
A value of 0 for this variable indicates the absence of a master’s technical
degree or a higher technical qualification (for example, PhD) of the founder.

Sales Capability of the Startup The sales capability of the startup is

measured as the number of customers/products offered at the time of primary
data collection (Ensley et al., 2003).

R&D Capability of the Startup The R&D capability of the startup is

measured as the average work experience of the R&D team measured in
number of years at the time of the data collection (Thornhill, 2006).

Financial Capability of the Startup Measured by a discrete dichotomous

variable that indicates whether or not the startup obtained funding external to
its founder’s and his family’s funds. This variable is labelled as ‘fin’ for the
analysis. A value of 1 for this variable indicates that the startup was funded
from external sources, and conversely, a value of 0 for this variable indicates
that the startup under consideration is not funded from external sources.

The external environment factors such as Industry sector, region/geography,

and policy are ensured to be controlled by the research design by limiting the
scope of study to one industry based out of one country that has the same
policies at macroeconomic levels.

4. Method of Analysis

In this study, we use Survival analysis as the method of analysis to examine

the stated objectives. Survival analysis deals with analyzing the time to the
incidence of an event. The survival model works on a set of assumptions,
primary ones being that when the observation of the data ends and the analysis
begins, the observed data set would typically have a combination of units in

67
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

such a fashion that the event in question has actually occurred for some units,
whereas the event may not have occurred for others (Aalen et al., 2008). The
key advantage of this model is that it helps the data analysts to deal with
missing information, often referred to as censored information. In the study, if
the startup has not yet achieved the product-market fit at the end of the data
collection phase, then this startup would be censored "on the right", that is, we
know that this particular startup’s survival time is known to exceed the time
duration between its formal creation and the closure of observation. Since
product-market fit is taken as proxy to measure startup survival, the above
scenario indicates that the startup has not yet achieved the survival milestone.
In survival analysis, the times at which certain events occur are assumed to
be realizations of some random process (Allison, 1995). So T, the time for an
event to occur for a particular observation, is a random variable having a
certain probability distribution. Different methods are used to model survival
data depending on the kind of distributions that the survival time T follows.
The survival function, which represents the unconditional probability of
surviving longer than “t” time units, has the following general form: S(t)
=Probability (T>t) = 1–F(t) where F(t) is the cumulative distribution function
of the random variable T, denoting time to failure (Chatterjee, 2010). The
focus of survival analysis would be to model the hazard rate h(t) which is
defined as h(t) = f(t)/S(t).
There are semi-parametric and parametric models to use with survival data.
The Cox Proportional Hazards Model (Cox, 1972) is the most widely used.
The Cox Proportional Hazards Model is popular because it does not require
one to make an assumption about the exact parametric form of the underlying
distribution of survival time. Also, in this model, hazards for two individuals
are proportional, with a proportionality constant that is independent of time.
However, in our study, since the independent variables change over time, and
that this change between the variables cannot be assumed as proportionally
constant, we use a parametric method of Accelerated Failure Time (AFT)
models for our analysis. These models are fitted based on the assumption that
survival times captured in the data follow certain well-known distributions
(Klein and Moeschberger, 1997).

V. Results and Discussion of Findings

The descriptive statistics for the variables that were used in the analyses are
presented in Table 1.

68
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Table 1 Descriptive statistics of the variables used for analysis

Min Max median mean Sd
stime 6 120 42 42.38 25.47
survst 1 2 2 1.73 0.45
fiexp 0 1 1 0.91 0.29
fsexp 0 1 0 0.49 0.51
fage 22 49 32 34.20 7.83
fedn 1 3 2 2.38 0.65
sales 1 1500 60 124.57 238.80
dev 0.5 7 2.5 2.89 1.81
Fin 0 1 0 0.33 0.48
Ol 0 1 1 0.67 0.48
dlogic 0 1 0 0.40 0.50

A standard estimator of the survival function in the presence of censoring is

the Kaplan-Meier Product Limit Estimator. This is a non-parametric method of
estimation. Plots of the Kaplan-Meier estimates of the survival function against
time provide a visual understanding of the survival function (Chatterjee, 2010).
For a visual inspection of the distribution of the survival time, the plot of the
survival function Ŝ(t) as estimated by the Kaplan-Meier estimator against time
is provided in Figure 2.

Figure 2 KM plot of survival probability against time in months

69
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

The full model containing all the independent variables and control variables
represented in R - an open source statistical software package is provided
below:

> model1=surv_data~fiexp+fsexp+fedn+fage+sales+dev+fin+ol+dlogic
> model1.step=stepAIC(survreg(model1))

For arriving at the most parsimonious model from the above full model,
Akaike’s Information Criterion (AIC) (Akaike, 1974) was used. AIC is a
numerical measure that weighs the likelihood of a model against its complexity.
The AIC of the AFT model is defined as: AIC = -2LL + 2 (c + a) where LL is
the logarithm of the model likelihood (log-likelihood), c is the number of
covariates and a is the number of ancillary parameters (Weibull distribution
has two parameters, λ and α, while exponential has only one parameter, λ). A
lower value of the AIC suggests a better model.
The appropriate distribution of survival times to be used for analysis is
determined by building AFT models for the above data using each of the
following distributions: Exponential, Weibull, Log-logistic and Log-normal.
The resulting AIC computed for each of the distributions used, which provides
the most optimal model containing the independent variables is presented in
Table 2.

Table 2 AIC computation for AFT models with different assumed distributions
Distribution Least AIC Value Optimium Model corresponding to least AIC Value

Exponential 343.24 surv_data ~ fsexp + fage + sales + fin + ol + dlogic

Weibull 310.03 surv_data~fsexp + fin + ol + dlogic

Log-Logistic 313.96 surv_data ~ fsexp + fin + ol + dlogic

surv_data ~ fiexp + fsexp + fage + dev + fin + ol +

Log-normal 318.58
dlogic

Since the Weibull model is found to have the lowest AIC for most of the
models we choose to use this as the standard distribution that we subsequently
discuss in this study. The output of the computation using the Weibull
distribution is provided in Table 3.

70
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Table 3 Results of the AFT optimum model execution

> optimum_model=surv_data~fsexp+fin+ol+dlogic
> summary(survreg(optimum_model, dist="weibull")
Call: survreg(formula=optimum_model, dist="weibull")
Value Std. Error Z P
(Intercept) 4.573 0.178 25.761 2.42E-146
fsexp1 -0.216 0.142 -1.521 1.28E-01
fin1 -0.481 0.149 -3.219 1.29E-03
ol1 -0.41 0.171 -2.4 1.64E-02
dlogic1 -0.12 0.157 -0.765 4.44E-01
Log(scale) -0.943 0.137 -6.902 5.14E-12
Scale=0.389
Weibull distribution
Loglik(model)=-149 Loglik(intercept only)=-157.5
Chisq=17.04 on 4 degrees of freedom, p=0.0019
Number of Newton-Raphson Iterations: 6
n=45

All the variables present in the most efficient model are highly significant
with very small p values, significant at 0.001 levels.
The results indicate that entrepreneur's prior startup experience, explorative
mode of learning transformation, causal decision-making of the entrepreneur
and availability of funding for the startup are the key factors that reduce the
time to survival of Indian high-tech startups. This study validates a portion of
the prior work related to the mode of learning transformation, impact of
learning outcomes and availability of funding on startups in the context of
emerging economies (Politis and Gabrielsson, 2005; Politis, 2008; Audretsch
and Lehmann, 2004; Suh et al., 2012; Kim and Ko, 2014). The empirical
results of this study pertaining to entrepreneurial decision-making preference
extend the knowledge in this field.
In this study, prior industry experience and prior startup experience were
introduced as two independent control variables. The results indicate that prior
startup experience will impact the survival time of a startup rather than prior
industry experience. The results indicate that prior startup experience will
accelerate the time to achieve survival by a factor of exp(-0.216)=0.8 times (i.e.
80% shorter survival time in comparison to the baseline). This is explainable,
specifically in the context of emerging economies, since the entrepreneur
would need to deal with an increased degree of liabilities of newness, given the
under developed infrastructure and environment support system that exists in

71
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

the region of operation. Having prior experience of starting up, dealing with
uncertainty, adds to the bundle of knowledge and therefore when the
entrepreneur embarks on the second venture, he would be much more prepared
to overcome the liabilities of newness in comparison to another individual,
who may have mere industry and domain knowledge, but ventures to start up
for the first time. These results also validate some of the earlier empirical
findings (Westhead et al., 2005; Politis, 2008).
Explorative mode of learning transformation relies on variance-seeking
learning. This mode of learning transformation has been shown to contribute to
huge success as well as failure (Politis and Gabrielsson, 2005). The results
indicate that explorative mode of learning will accelerate the time to achieve
survival by a factor of exp(-0.41)=0.66 times (i.e. 66% shorter survival time in
comparison to the baseline). This study revalidates the findings obtained in
prior literature this studied exploratory learning under different contexts.
Ucbasaran and Westhead (2002) stated that opportunity-seeking entrepreneurs,
who continuously scan the environment for new entrepreneurial opportunities
often strive for variation with the goal to learn something new. Politis and
Gabrielsson (2005) observe that entrepreneurs who are highly explorative are
alert and would become more effective in identifying and acting on
entrepreneurial opportunities.
Causal decision-making helps in reducing the liability of newness as it
focuses on converging entrepreneurial actions towards mean performance and
repeatability. The decision-making based on past data and experience tries to
bring structure and direction to the overall activities of the startup, which
usually operates in a volatile environment. Hence, causal decision-making by
the entrepreneurs reduces the risks that arise out of dealing with liability of
newness and therefore help in reducing time to survival of the startup. The
results indicate that causal-decision making will accelerate the time to achieve
survival by a factor of exp(-0.12)=0.89 times (i.e. 89% shorter survival time in
comparison to the baseline).
Funding has long been researched and established as a key factor that
contributes to the survival of a startup. The results indicate that funding will
accelerate the time to achieve survival by a factor of exp(-0.481)=0.62 times
(i.e. 62% shorter survival time in comparison to the baseline). Audretsch and
Lehmann (2004) established funding as being a key factor in influencing the
survival of high-tech startups. Funding of the startup helps the entrepreneur in
multiple ways. It frees the entrepreneur to focus his energies and abilities to
exploit the entrepreneurial opportunity, and also provides additional cushion to
react and deal with uncertain circumstances that a startup might face in its
early stages.

72
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

VI. Summary and Inferences

This study contributes to the existing body of knowledge on entrepreneurial

learning and high-tech startup survival. At the outset, by taking an end-to-end
examination approach, this study investigated the linkage between the mode of
transformation of entrepreneurial learning into outcomes and the subsequent
impact of these learning outcomes in enhancing the survival of high-tech
startups in India. Two modes of transformation of entrepreneurial learning -
explorative and exploitative - and two preferences of entrepreneurial decision-
making (learning outcome) - effectual and causal - were examined for their
impact and influence on startup survival. This initial attempt to understand the
entire process is a contribution to the literature and research on entrepreneurial
learning and entrepreneurial decision-making domains.
The study uses data collected from 45 high-tech startups operating across
different cities in India. Survival Analysis of the data using Accelerated Failure
Time models is conducted to determine which mode of learning transformation
and what type of entrepreneurial decision making preference have a major
influence on Indian high-tech startups survival. The results indicate that
entrepreneur's prior startup experience, explorative mode of learning
transformation, causal decision-making of the entrepreneur and availability of
funding for the startup are the key factors that reduce the time to survival of
Indian high-tech startups. The results of this study provide insights on how the
entrepreneur, through exploratory learning transformation, will increase his
skills in recognizing opportunity (Shane and Venkataraman, 2000) and, by
leveraging causal decision-making preference, would be able to cope with the
liabilities of newness (Stinchcombe, 1965). The findings provide a better
understanding of the mode of learning transformation and dominant decision-
making preferences of startup entrepreneurs operating in emerging economies
such as India. They also provide key insights on how these factors impact the
startup survival in this region.
This paper validates a portion of prior work related to the mode of learning
transformation and availability of funding for startups in the context of
emerging economies. The empirical results of this study pertaining to
entrepreneurial decision-making preference expand the knowledge in this field.
For high-tech startup entrepreneurs in emerging economies, it provides insights
on the factors they need to focus on enhancing their chances of survival. For
the policy-makers, investors and practitioners and other stakeholders who are
focused on emerging economies, the outcomes of this study provides insights
on the type of factors that need to be kept in mind to create a vibrant startup
ecosystem in the region. Next, the study explains how the variance in learning

73
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

outcomes, in this case by conscious choice of a preferred decision-making, will

help reduce or advance the time to survival of the entrepreneur’s startup.
Finally, for entrepreneurs contemplating on setting up high-tech startups in
emerging economies, it provides insights on the factors they need to focus on
enhancing their chances of survival and success. For policy-makers, investors
and practitioners focusing on emerging economies, it reveals the type and kind
of micro factors that should be examined to enable a vibrant startup ecosystem
in the region.
However, this study has certain limitations for generalization. There is scope
for further insights and investigation into the micro aspects of additional
factors impacting the learning transformation of entrepreneurs, as well as on
factors that influence the preference of decision-making of entrepreneurs. For
example, one could further examine if there is any preference of learning
transformation of experience into knowledge between novice and experienced
entrepreneurs. Next, this study examines only the dominant mode of learning
transformation and preference to decision-making among entrepreneurs. In
reality, entrepreneurs, based on the context, use both modes of learning
transformation and decision-making. Further, most variables considered in this
study are binary in nature, thereby limiting the ability to study the correlations,
and also the ability to study other dimensions of decision making such as
trade-off between effectiveness and risk.
Lastly, Survival analysis as a method needs to evolve substantially to
ensure higher reliability of the results. Since the initiation and termination of
research observations are essentially artificial, there is always an instance in
which it is impossible to confirm whether or not survival is possible. Survival
analysis is very complicated due to the complexity of the outcome variables
and incomplete censoring, and there are many problems that have not yet been
solved statistically.

Acknowledgements
The author acknowledges the comments and suggestions received from the
anonymous reviewers that have helped enhance the quality of the paper.

74
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

References

Aalen, O.O., Borgan, Ø . and Gjessing, H.K. (2008) Survival and Event History
Analysis: A Process Point of View, New York: Springer-Verlag, .
Akaike, H. (1974) A new look at the statistical model identification, IEEE Transaction
on Automatic Control AC-19, 716-723.
Allison, P.D. (1995) Survival Analysis Using SAS® : A Practical Guide, Cary, NC:
SAS Institute Inc.
Argyris, C. and Schön, D.A. (1978) Organizational Learning: A Theory of Action
Perspective, Reading, MA: Addison-Wesley.
Audretsch, D.B. and Lehmann, E.E. (2004) Financing high-tech growth: the role of
banks and venture capitalists, Schmalenbach Business Review, 56, 340-357.
Balasubramanian, N. (2011) New plant venture performance differences among
incumbent, diversifying, and entrepreneurial firms: the impact of industry learning
intensity, Management Science, 57, 549-565.
Boeker, W. (1988) Strategic change: the effects of founding and history, Academy of
Management Journal, 32(3), 489-515.
Brettel, M., Mauer, R., Engelen, A. and Kupper, D. (2012) Corporate effectuation:
entrepreneurial action and its impact on R&D project performance, Journal of
Business Venturing, 27, 167-184.
Brush, C.G., Green, P.G. and Hart, M.M. (2001) From initial idea to unique advantage:
the entrepreneurial challenge of constructing a resource base, Academy of
Management Executive, 15, 64-78.
Chaston, I., Badger, B. and Sadler-Smith, E. (2001) Organizational learning: an
empirical assessment of process in small U.K. manufacturing firms, Journal of Small
Business Management, 39, 139-151.
Chatterjee, D. (2010) Studies on some aspects of liquidity of stocks: limit order
executions in the Indian stock market, Ph.D thesis, Department of Management
Studies, Indian Institute of Science.
Cohen, W.M., Levinthal, D.A. (1990) Absorptive capacity: a new perspective on
learning and innovation, Administrative Science Quarterly, 35, 128-152.
Cope, J. (2003) Entrepreneurial learning and critical reflection: discontinuous events as
triggers for higher-level learning, Management Learning, 34, 429-450.
Covin, J.G., Green, K.M. and Slevin, D.P. (2006) Strategic process effects on the
entrepreneurial orientation-sales growth rate relationship, Entrepreneurship Theory
and Practice, 30, 57-81.
Covin, J.G., Garrett, R.P., Gupta, J.P., Kuratko, D.F. and Shepherd, D.A. (2016) The
Interdependence of Planning and Learning Among Internal Corporate Ven-
tures, Entrepreneurship Theory and Practice.
Cox, D.R. (1972) Regression models and life-tables with discussion, Journal of the
Royal Statistical Society: Series B, 34, 187-220
Douglas, D. (2005) The human complexities of entrepreneurial decision making: a
grounded case considered, International Journal of Entrepreneurial Behaviour and
Research, 11(6), 422-435.

75
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Ensley, M.D., Pearson, A. and Pearce, C.L. (2003) Top management team process,
shared leadership and new venture performance: a theoretical model and research
agenda, Human Resource Management Review, 13, 329-346.
Furdas, M. and Kohn, K. (2011) Why is startup survival lower among necessity
entrepreneurs? a decomposition approach, Preliminary Version, April 2011.
Gabrielsson, J. and Politis, D. (2011) Career motives and entrepreneurial decision-
making: examining preferences for causal and effectual logics in the early stage of
new ventures, Small Business Economics, 36, 281-298.
Haber, S. and Reichel, A. (2007) The cumulative nature of the entrepreneurial process:
the contribution of human capital, planning and environment resources to small
venture performance, Journal of Business Venturing 22(1), 119-145.
Huber, G.P. (1991) Organizational learning: the contributing processes and the
literatures, Organization Science, 2, 88-115.
Kim, C.H. and Ko, C.R. (2014) Success and failure factors of technology commer-
cialization: a Korean case, Asian Journal of Innovation and Policy, 3(1), 25-49.
Kim, Y.J. and Shin, S.J. (2017) What causes technology commercialization to succeed
or fail after transfer from public research organizations, Asian Journal of Innovation
and Policy, 6(1), 23-44.
Kirchhoff, B.A. (1994) Entrepreneurship and Dynamic Capitalism: The Economics of
Business Firm Formation and Growth, Westport, Conn: Praeger.
Kirchhoff, B.A. and Spencer, A. (2008) New high tech firm contributions to economic
growth, Proceedings of International Council for Small Business World Conference,
Halifax, Canada: Nova Scotia.
Klein, J.P. and Moeschberger, M.L. (1997) Survival Analysis, Techniques for
Censored and Truncated Data, New York: Springer-Verlag, 502.
Kolb, D.A. (1984) Experiential Learning: Experience as the Source of Learning and
Development, Englewood Cliffs, NJ: Prentice Hall.
Lant, T.K. and Mezias, S.J. (1990) Managing discontinuous change: a simulation study
of organizational learning and entrepreneurship, Strategic Management Journal, 11,
147-179.
Lamont, L.M. (1972) What entrepreneurs learn from experience, Journal of Small
Business Management, 10, 36-41.
Lévesque, M., Minniti, M. and Shepherd, D. (2009) Entrepreneurs’ decisions on timing
of entry: learning from participation and from the experiences of others,
Entrepreneurship Theory and Practice, 33, 547-570.
Levinthal, D.A. and March, J. (1993) The myopia of learning, Strategic Management
Journal, 14, 95-112.
March, J.G. (1991) Exploration and exploitation in organizational learning,
Organization Science, 2(1), 71-87.
Markman, G.D. and Baron, R.A. (2003) Person-entrepreneurship fit: why some people
are more successful as entrepreneurs than others, Human Resource Management
Review, 13, 281-301.
McGrath, R.G. (2001) Exploratory learning, innovative capacity, and managerial
oversight, Academy of Management Journal, 44, 118-131.
Mian, S., Lamine, W. and Fayolle, A. (2016) Technology business incubation: an
overview of the state of knowledge, Technovation, 50, 1-12.

76
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Minniti, M. and Bygrave, W. (2001) A dynamic model of entrepreneurial learning,

Entrepreneurship Theory and Practice, 25(3), 5-16.
Mintzberg, H., Waters, J.A. (1982) Tracking strategy in an entrepreneurial firm.
Academy of Management Journal, 25(3), 465-499.
Politis, D. (2008) Does prior start-up experience matter for entrepreneurs’ learning?,
Journal of Small Business and Enterprise Development, 15, 472-489.
Politis, D. and Gabrielsson, J. (2005) Exploring the role of experience in the process of
entrepreneurial learning, Working Paper Series, Lund Institute of Economic
Research, 2005/1.
Rae, D. and Carswell, M. (2001) Towards a conceptual understanding of entre-
preneurial learning, Journal of Small Business and Enterprise Development, 8, 150-
158.
Rae, D. (2005) Entrepreneurial learning: a narrative-based conceptual model, Journal of
Small Business and Enterprise Development, 12, 323-335.
Rerup, C. (2005) Learning from past experience: footnotes on mindfulness and habitual
entrepreneurship, Scandinavian Journal of Management, 21, 451-72.
Reuber, R.A., Dyke, L.S. and Fischer, E.M. (1990) Experiential acquired knowledge
and entrepreneurial venture success, Academy of Management Best Paper
Proceedings, 69-73.
Reuber, R.A. and Fischer, E.M. (1994) Entrepreneurs’ experience, expertise, and the
performance of technology-based firms, IEEE Transactions on Engineering
Management, 41(4), 365-374.
Reuber, R.A. and Fischer, E.M. (1999) Understanding the consequences of founders’
experience, Journal of Small Business Management, 37(2), 30-45.
Ronstadt, R. (1988) The corridor principle, Journal of Business Venturing, 3, 31-40.
Sarasvathy, S.D. (2001) Causation and effectuation: toward a theoretical shift from
economic inevitability to entrepreneurial contingency, Academy of Management
Review, 26(2), 243-263.
Sarasvathy, S.D. and Dew, N. (2005) New market creation through transformation,
Journal of Evolutionary Economics, 15(5), 533-565.
Sarasvathy, S.D. (2008) Effectuation: Elements of Entrepreneurial Expertise,
Northhampton, MA and Cheltenham: Edward Elgar.
Sardana, D. and Scott-Kemmis, D. (2010) Who learns what? a study based on
entrepreneurs from biotechnology new ventures, Journal of Small Business
Management, 48, 441-468.
Shane, S. and Venkataraman, S. (2000) The promise of entrepreneurship as a field of
research, Academy of Management Review, 25, 217-226.
Shepherd, D.A., Douglas, E.J. and Shanley, M. (2000) New venture survival: ignorance,
external shocks, and risk reduction strategies, Journal of Business Venturing, 15,
393-410.
Stinchcombe, A.L. (1965) Social Structure and Organizations, In: James, G. (ed.),
Handbook of Organizations, Chicago: Rand McNally, 142-193.
Suh, M.K., Park, J.O. and Park, B.S. (2017) Determinants of R&D commercialization
by SMEs after technology transfer, Asian Journal of Innovationand Policy, 6(1), 45-
57.

77
 Asian Journal of Innovation and Policy (2018) 7.1:055-078

Suh, S.H., Ko, J.O. and Lee, S.Y. (2012) An empirical study on the success factors of
inter-firm alliances for new product development: with a focus on the SMEs in
Korea, Asian Journal of Innovation and Policy, 1(1), 71-91.
Sullivan, R. (2000) Entrepreneurial learning and mentoring, International Journal of
Entrepreneurial Behaviour and Research, 6, 160-175.
Thornhill, S. (2006) Knowledge, innovation and firm performance in high and low
technology regimes, Journal of Business Venturing, 21(5), 687-703.
Ucbasaran, D. and Westhead, P. (2002) Does entrepreneurial experience influence
opportunity identification?, In: Bygrave, W.D. et al., (eds.), Frontiers of Entre-
preneurship Research, Wellesley, MA: Babson College.
Vohora, A., Wright, M. and Lockett, A. (2004) Critical junctures in the development of
university high-tech spin-out companies, Research Policy, 33(1), 147-175.
Wang, C.L. (2008) Entrepreneurial orientation, learning orientation, and firm
performance, Entrepreneurship Theory and Practice, 32, 635-656.
Wang, C.L. and Chugh, H. (2014) Entrepreneurial learning: past research and future
challenges, International Journal of Management Reviews, 16, 24-61.
Watson, W., Steward, W. and BarNir, A. (2003) The effects of human capital,
organizational demography, and interpersonal processes on venture partner
perceptions of firm profit and growth, Journal of Business Venturing 18(2), 145-164.
Westhead, P., Ucbasaran, D. and Wright, M. (2005) Decisions, actions, and
performance: do novice, serial, portfolio entrepreneurs differ?, Journal of Small
Business Management, 43, 393-417.
Wiklund, J., Patzelt, H. and Shepherd, D.A. (2009) Building an integrative model of
small business growth, Small Business Economics, 32, 351-374.
Williams, M.K. and Donnellon, A. (2014) Personalizing entrepreneurial learning: a
pedagogy for facilitating the know why, Entrepreneurship Research Journal, 4(2),
167-204.
Wiltbank, R., Read, S., Dew, N. and Sarasvathy, S. (2006) What to do next? the case
for non-predictive strategy, Strategic Management Journal, 27, 981-998.
Young, J.E., Sexton, D.L. (2003) What makes entrepreneurs learn and how do they do
it?, Journal of Entrepreneurship, 12, 155-182.

78